麻豆传媒

Endorsing COVID-19 vaccine misinformation, lack of trust in physicians or scientists, and conspiracy-mindedness were linked to the use of non-evidence-based treatments for COVID-19, an internet-based survey study of U.S. adults showed.

Of more than 13,000 survey respondents who reported prior COVID infection, those who endorsed at least one item of COVID-19 vaccine misinformation were more likely to have received ivermectin or hydroxychloroquine (adjusted OR 2.86, 95% CI 2.28-3.58), reported Roy H. Perlis, MD, MSc, of Massachusetts General Hospital and Harvard Medical School in Boston, and colleagues.

Also more likely to receive non-evidence-based medication were those reporting trust in social media (aOR 2.39, 95% CI 2.00-2.87) and in Donald Trump (aOR 2.97, 95% CI 2.34-3.78), and those with greater scores on the American Conspiracy Thinking Scale (aOR 1.10, 95% CI 1.07-1.13), they noted in .

"In aggregate, our survey results suggest that the potential harms of misinformation may extend to prescription of ineffective and potentially toxic treatments, rather than simply avoiding health-promoting behaviors, such as vaccination," Perlis and team wrote.

Conversely, people who reported trusting physicians and hospitals (aOR 0.74, 95% CI 0.56-0.98) and scientists (aOR 0.63, 95% CI 0.51-0.79) were less likely to seek non-evidence-based treatments.

Perlis told 麻豆传媒 that "about 6% of adults had taken either ivermectin or hydroxychloroquine -- and sometimes both. By comparison, 7.3% of adults had taken an FDA-approved antiviral medication."

"When we accounted for demographic differences, we didn't find differences between Democrats or Republicans in using these non-evidence-based medications," Perlis said. "I think we might have expected to see differences by political affiliation, since so many aspects of the pandemic became politicized. But I think the key point is that misinformation matters, and trust matters, far more than political party."

The authors noted that "[t]his finding counters many narratives that focus on how partisan divides drive gaps in COVID-19-related behaviors. This distinction is important insofar as altering one's party identification may be an unrealistic bar, whereas combating misinformation and trying to build institutional trust may be more feasible."

Perlis said he and his team are looking further at the survey data "to try to inform public health strategies in responding to COVID, or the next pandemic." Improving trust in doctors and scientists is "critical if we are to make progress in public health," he added.

For this study, the researchers used data from the COVID States Project, which Perlis described as "a large survey that started in the spring of 2020 in an effort to understand people's attitudes and behavior during the pandemic." Every 4 to 8 weeks, the project conducts a 50-state internet survey of people 18 and older. The survey applies representative state-based quotas for race, ethnicity, age, and gender. The data for this study were from wave 26, which encompassed Dec. 22, 2022 through Jan. 16, 2023.

In total, 13,438 respondents reported having had COVID-19 before. Mean age was 42.7, 68.1% were women, 72.5% were white, and 37.8% had a college degree. About 35% reported being a Democrat, while 28.8% reported being a Republican and 36.5% reported being an Independent or "other."

The survey explicitly asked about both hydroxychloroquine and ivermectin, as well as FDA-approved antivirals, including molnupiravir (Lagevrio) and nirmatrelvir-ritonavir (Paxlovid).

Of the respondents, 3.9% reported use of hydroxychloroquine, 3.3% reported use of ivermectin, 2% reported use of molnupiravir, and 6.1% reported use of nirmatrelvir-ritonavir.

Further survey questions assessed whether respondents could identify COVID-19 misinformation, how much they trusted institutions, and their level of conspiracy-mindedness, using the 4-item American Conspiracy Thinking Scale.

The authors noted a few limitations, including that they could not estimate response rates to the survey because of its opt-in internet nonprobability design. The study was also cross-sectional, meaning they could not infer causality between misinformation and mistrust and pursuing non-evidence-based treatments. Plus, data on all diagnoses and treatment were self-reported, which doesn't have the same level of certainty as data from electronic health records.

Comment